Portal de Periódicos da CAPES

Making It Personal: Neoantigen Vaccines in Metastatic Melanoma

2017; Cell Press; Volume: 47; Issue: 2 Linguagem: Inglês

10.1016/j.immuni.2017.08.001

1097-4180

Matthew D. Hellmann, Alexandra Snyder,

Cancer Immunotherapy and Biomarkers

Somatic mutations in cancer can be translated into peptides, termed neoantigens, which can be recognized by the immune system as "foreign" epitopes. Two recent studies in Nature (Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar, Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar) examine the effects of neoantigen vaccines on patients with stage III or IV melanoma and demonstrate immunogenicity and intriguing clinical safety and efficacy data in phase I studies. Somatic mutations in cancer can be translated into peptides, termed neoantigens, which can be recognized by the immune system as "foreign" epitopes. Two recent studies in Nature (Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar, Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar) examine the effects of neoantigen vaccines on patients with stage III or IV melanoma and demonstrate immunogenicity and intriguing clinical safety and efficacy data in phase I studies. The concept that tumor rejection could be mediated by tumor-specific features dates conceptually to Paul Ehrlich around the turn of the 20th century and empirically to experiments by several investigators who showed that immunocompetent mice reject tumors to which they have previously been exposed upon rechallenge, suggesting an immune-mediated response against tumor-specific antigens (Prehn and Main, 1957Prehn R.T. Main J.M. J. Natl. Cancer Inst. 1957; 18: 769-778PubMed Google Scholar). More recent studies of patients with responses to checkpoint blockade therapies have documented neoantigen-specific T cell responses in responding patients (Le et al., 2017Le D.T. Durham J.N. Smith K.N. Wang H. Bartlett B.R. Aulakh L.K. Lu S. Kemberling H. Wilt C. Luber B.S. et al.Science. 2017; 357: 409-413Crossref PubMed Scopus (3771) Google Scholar, Rizvi et al., 2015Rizvi N.A. Hellmann M.D. Snyder A. Kvistborg P. Makarov V. Havel J.J. Lee W. Yuan J. Wong P. Ho T.S. et al.Science. 2015; 348: 124-128Crossref PubMed Scopus (5531) Google Scholar, Snyder et al., 2014Snyder A. Makarov V. Merghoub T. Yuan J. Zaretsky J.M. Desrichard A. Walsh L.A. Postow M.A. Wong P. Ho T.S. et al.N. Engl. J. Med. 2014; 371: 2189-2199Crossref PubMed Scopus (3028) Google Scholar, van Rooij et al., 2013van Rooij N. van Buuren M.M. Philips D. Velds A. Toebes M. Heemskerk B. van Dijk L.J. Behjati S. Hilkmann H. El Atmioui D. et al.J. Clin. Oncol. 2013; 31: e439-e442Crossref PubMed Scopus (653) Google Scholar) and provide evidence for tumor sculpting by neoantigen-reactive T cells (Verdegaal et al., 2016Verdegaal E.M. de Miranda N.F. Visser M. Harryvan T. van Buuren M.M. Andersen R.S. Hadrup S.R. van der Minne C.E. Schotte R. Spits H. et al.Nature. 2016; 536: 91-95Crossref PubMed Scopus (308) Google Scholar). Preclinical data by several groups suggested efficacy for a neoantigen vaccine in immune-competent mouse models (Gubin et al., 2014Gubin M.M. Zhang X. Schuster H. Caron E. Ward J.P. Noguchi T. Ivanova Y. Hundal J. Arthur C.D. Krebber W.J. et al.Nature. 2014; 515: 577-581Crossref PubMed Scopus (1433) Google Scholar). Furthermore, multiple studies have demonstrated that expanded autologous tumor-infiltrating lymphocyte are reactive against tumor neoantigens and can cause tumor regressions in some patients (Tran et al., 2016Tran E. Robbins P.F. Lu Y.C. Prickett T.D. Gartner J.J. Jia L. Pasetto A. Zheng Z. Ray S. Groh E.M. et al.N. Engl. J. Med. 2016; 375: 2255-2262Crossref PubMed Scopus (739) Google Scholar). It has also been shown that naive T cells from non-cancer patients could react against tumor neoantigens, providing evidence that de novo anti-neoantigen responses could be elicited (Strønen et al., 2016Strønen E. Toebes M. Kelderman S. van Buuren M.M. Yang W. van Rooij N. Donia M. Böschen M.L. Lund-Johansen F. Olweus J. Schumacher T.N. Science. 2016; 352: 1337-1341Crossref PubMed Scopus (298) Google Scholar). These studies raise the question of whether anti-neoantigen T cell responses be induced de novo or augmented by vaccination with predicted neoantigens to treat patients with cancer. Two studies recently published in Nature by Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar and Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar suggest that the answer is "yes." These studies provide evidence that vaccines against tumor neoantigens—immunogenic peptides resulting from somatic mutations—may be immunologically and therapeutically effective. Both studies come to harmonious conclusions in similar patient populations, although they use different vaccination methods. In both studies, tumors from each patient underwent whole-exome and transcriptome sequencing followed by human leukocyte antigen (HLA)-specific neoantigen prediction. In Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar, 6 patients with resected stage III-IV melanoma received personalized vaccine with up to 20 long peptide neoantigens per patient (toll-like receptor 3 [TLR3] and melanoma differentiation-associated protein 5 [MDA-5] poly-ICLC were also co-administered as adjuvants). In Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar, a personalized RNA-based vaccine was used to treat 13 patients with stage III-IV melanoma. Although Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar prioritizes HLA class I prediction and Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar prioritizes HLA class II prediction, both use long peptide vaccines, in essence leaving it to the cell's peptide processing and antigen-presenting machinery to determine which portion of the peptide is presented for T cell receptor (TCR) recognition. Both studies find a higher proportion of CD4+ than CD8+ T cell responses, and some peptides elicited both CD4+ and CD8+ T cell responses. Importantly, both studies also document a broadening of the neoantigens recognized after vaccination. Specifically, in Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar, expression data suggest that the CD4+ T cells studied in one experiment transitioned from a naive to effector to memory phenotype. In Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar, one-third of peptides strengthened existing anti-neoepitope responses while two-thirds developed de novo. In other words, both studies suggest that in addition to preexisting immunity to many of the predicted neoepitopes, naive cells were activated by the vaccine. In addition to the exciting immunologic results, the clinical results in both studies are intriguing despite the small sample sizes. None of the four stage III patients in the Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar study showed melanoma recurrence, while the two patients with metastatic disease who did relapse later experienced complete responses to anti-programmed cell death protein 1 (PD-1) therapy. In the Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar study, among eight patients with no detectable disease at the time of vaccination, most experienced prolonged disease-free survival. And perhaps most promising, among five patients with evident metastatic disease at the time of vaccination, two experienced objective responses that were attributable to the vaccine alone. The idea that personalized immune manipulation could cure patients who are at high risk of progression and death is tantalizing. The technical expertise and infrastructural coordination needed to analyze, generate, and deliver a personalized cancer vaccine is truly monumental; both groups have achieved remarkable feats in these reports, representing critical steps forward for the future success of this therapeutic approach. Nevertheless, as with any early trial result, the challenges ahead are substantial. Major residual questions, among others, are listed in Figure 1. What is the ideal clinical setting in which to apply a vaccine strategy? In the adjuvant setting, when patients have no radiographic evidence of disease after surgical resection of the primary tumor, there may be an opportunity for vaccination to re-stimulate or initiate anti-tumor immunity to optimize eradication of residual micrometastases. However, it is possible that antigen stimulation will be too weak in the absence of a primary tumor, or vaccination based on a tumor primary could be ineffective against clonally distinct metastases. When patients have metastatic disease, vaccination could potentially enhance antigen stimulation of intact tumor; however, if the tumor is present, how much will adding tumor-associated peptides contribute? In vaccinations against infectious diseases, one does not generally vaccinate actively infected patients as the pathogen is already providing the best possible antigens. To our knowledge, the follow-up studies of these neoantigen vaccine approaches will explore the metastatic setting, some in conjunction with checkpoint blockade and/or adjuvants, which will help evaluate the use of vaccines in patients with metastatic disease. Another pertinent clinical question is whether neoantigen vaccines will be effective in tumors with lower neoantigen burden, a population that represents an area of unmet need within immunotherapy. Practical aspects of personalized vaccine manufacture require ongoing optimization. The issue of routine feasibility and expedited pace is particularly relevant in the metastatic setting, when time is of the essence. In the Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar study, of ten patients enrolled, eight had peptide vaccines successfully synthesized and six were actually vaccinated. The fastest time to vaccination was 3 months, with a median of 4.5 months. The time from tissue acquisition to vaccine delivery in the Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar study is not entirely clear, but the time from selection of mutations to vaccine release alone (seemingly not inclusive of tissue acquisition, sequencing, and initial analysis steps) ranged from 89 to 160 days. Certainly this time to vaccine delivery will need to be substantially shortened for this strategy to become feasible in patients with metastatic disease. Beyond manufacturing, optimization is also critically needed for prediction of immunologically relevant neoantigens. In the Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar study, for example, despite the vaccine being based on MHC class I prediction, CD4+ T cell responses were even more common than CD8+ responses to vaccinated neoantigens. Unbiased methods to test T cell reactivity against every mutation in a tumor (removing predictions about peptide processing, MHC affinity, similarity to self antigens, and TCR recognition) will be helpful to challenge current assumptions and iteratively inform and refine what features define the antigenicity and immunogenicity of neoantigens. The studies by Ott et al., 2017Ott P.A. Hu Z. Keskin D.B. Shukla S.A. Sun J. Bozym D.J. Zhang W. Luoma A. Giobbie-Hurder A. Peter L. et al.Nature. 2017; 547: 217-221Crossref PubMed Scopus (1533) Google Scholar and Sahin et al., 2017Sahin U. Derhovanessian E. Miller M. Kloke B.P. Simon P. Löwer M. Bukur V. Tadmor A.D. Luxemburger U. Schrörs B. et al.Nature. 2017; 547: 222-226Crossref PubMed Scopus (11) Google Scholar are independent and differ in their approach, yet reach similar conclusions that certainly demand further exploration and expansion. First, both demonstrate that neoantigen vaccines can be successfully synthesized and safely administered to patients with melanoma. Second, they show that these vaccines, whether peptide or RNA based, induce polyfunctional, neoantigen-specific CD4+ and CD8+ T cell responses. Third, vaccination can elicit de novo neoantigen-specific responses from naive T cells in addition to augmenting pre-existing immunity. Fourth, and crucially, these reports provide preliminary clinical evidence that personalized neoantigen vaccination can direct therapeutic anti-tumor immunity. As technological improvements accelerate the pace with which neoantigen vaccines can be synthesized, later-phase studies may demonstrate efficacy in melanoma and beyond, with potential enhancement by administration of immune checkpoint blockade. M.D.H. was formerly a consultant for Neon Therapeutics and currently is a consultant for Genentech, BMS, Merck, AstraZeneca, Janssen, and Novartis. A.S. was formerly a consultant for Neon Therapeutics and is currently an employee of Adaptive Biotechnologies.